Abstract
BACKGROUND: Gingival phenotype (GP) significantly influences periodontal health and treatment outcomes. Traditional methods for assessing GP, using gingival thickness (GT) alone, may lack sufficient accuracy for reliable GP classification. The present study aims to introduce and validate a novel non-invasive digital GP assessment measuring digital GT (dGT) and digital keratinized tissue width (dKTW). The primary objective is to assess the diagnostic performance of digital gingival phenotype (dGP) in distinguishing between thick and thin phenotypes. METHODS: This prospective, cross-sectional study was conducted at the Periodontology Department of Istanbul University-Cerrahpaşa, Turkey, from October to December 2024. Participants were included if they had all maxillary and mandibular anterior teeth present. Exclusion criteria included factors that could affect periodontal tissues, such as clinical attachment loss, systemic diseases (e.g., diabetes), gingival enlargement or recession, smoking, medications causing gingival hyperplasia, and melanin pigmentation. All subjects were screened for eligibility by S.K. prior to enrollment, with participants enrolled when S.K. was present at the periodontology department for preliminary examinations using a convenience sampling approach. KTW was assessed using clinical (cKTW), digital (dKTW), and rounded methods. cKTW and dKTW measured the distance between the gingival margin and mucogingival junction, while rounded KTW was calculated by rounding dKTW to the nearest whole number. GT was measured digitally in millimeters from the gingival margin level. GP was evaluated clinically (cGP) with a color-coded periodontal probe and digitally (dGP) by multiplying dKTW and dGT measurements. cKTW, dKTW, rounded KTW, dGT, and dGP are index tests, with cGP serving as the reference standard. The diagnostic accuracy of each method was evaluated using Receiver Operating Characteristic (ROC) analysis. RESULTS: Out of 348 participants, 31 met the inclusion criteria. Since each participant's 12 teeth were evaluated, a total of 372 teeth were included in the study. The area under the curve (AUC) values and 95% confidence intervals (CI) for each method were as follows: dGT: 0.628 (95% CI: 0.570-0.687), cKTW: 0.730 (95% CI: 0.677-0.782), dKTW: 0.714 (95% CI: 0.661-0.767), rounded KTW: 0.710 (95% CI: 0.657-0.763), and dGP: 0.734 (95% CI: 0.683-0.785). The dGP model exhibited the highest diagnostic accuracy, while the dGT model showed the lowest. CONCLUSIONS: The findings suggest that the digital gingival phenotype assessment provides superior diagnostic accuracy compared to other methods, achieving the highest AUC value. This demonstrates its efficacy in classifying GP and offers a reliable and accurate alternative to traditional clinical techniques for GP classification. REGISTRATION: No trial registration was performed, as no invasive procedures were conducted in this study.